ZOOM LINKMeeting ID: 981 1878 3645Passcode: 219869 Abstract: A tremendous number of quantitative measurements have been made on biological cellular pathways, contributing to a massive body of knowledge about how living things work. Yet, we are scarcely able to quantitatively predict the behavior of whole biological cellular systems from the measures of their constituent molecular parts and cell populations. Furthermore, current technologies cannot directly measure many biological quantities of interest. To aid in addressing these gaps, we contribute dynamical models of cellular pathways as analog electrical circuits – suitable for analog and probabilistic computing in digitally programmable chips. They may also be represented and simulated in deterministic and stochastic ordinary differential equations (ODEs) on traditional digital computers. Specifically, we develop a model of SARS-CoV-2 infection that fits biological data; explains the effects of age, sex, viral variant, and treatment type on infection dynamics; and enables drug cocktail formulation via circuit design. Then, we present six fundamental behaviors of immune-pathogen feedback circuits that emerge as we vary immune vs. pathogen parameters in a parameter-outcome phase plane plot. We demonstrate an application of Dijkstra’s path-finding algorithm that selects optimal dose amounts, dose schedules, and drug cocktails for treatment regimen design. The algorithm helps place us in the “cure outcome” region of our prior phase plane plots. ATP/energy metabolism is known to be important in cellular stochastics and antibiotic persistence. We create an analog circuit model of E. coli ATP metabolism that fits experimental data to infer ATP production and consumption rates across growth stages. We also fit a novel model of antibiotic persister cell dynamics to biphasic killing curve data from antibiotic-treated E. coli cultures in order to measure the antibiotic-induced persister formation rate. Our work demonstrates the many ways in which analog electrical circuit models can be beneficially used to predict, analyze, design, simulate, and repair biological systems. Thesis Committee: Rahul Sarpeshkar (chair), Margaret Ackerman, Daniel Schultz, Edward Stites (Yale University)